|Publication number||US4592032 A|
|Application number||US 06/614,348|
|Publication date||May 27, 1986|
|Filing date||May 24, 1984|
|Priority date||Feb 5, 1981|
|Publication number||06614348, 614348, US 4592032 A, US 4592032A, US-A-4592032, US4592032 A, US4592032A|
|Inventors||Gary M. Ruckgaber|
|Original Assignee||Standard Oil Company (Indiana)|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (10), Non-Patent Citations (10), Referenced by (8), Classifications (9), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This is a continuation of copending application Ser. No. 231,699, filed Feb. 5, 1981, now abandoned.
The invention relates to methods of processing geophysical data and to displays of the processed data. In one aspect, the invention relates to the processing and display of data obtained by seismic methods.
In seismic prospecting, an acoustic or seismic wave energy source is positioned either on the surface or in the subsurface of the earth, or is towed a predetermined distance beneath the surface of a body of water. A seismic source signal of some preselected waveshape and energy content is generated by the seismic energy source. The energy generated by the source penetrates through the layers of material comprising the subterranean portion of the earth in the region being prospected. Since the energy propagates at different speeds in each layer of material, refractions, reflections, and diffractions of the seismic energy occur at each layer. These energy reflections, refractions, and diffractions cause secondary seismic energy signals to return toward the surface of the earth. At the surface, the returning seismic signals are detected by a plurality of geophones, hydrophones, or seismic detectors spaced in a geometric array. The detectors generate electrical signals in analog or digital form which are representative of the amplitude of the reflected, refracted, and diffracted seismic energy signals arriving at the detectors.
The seismic signals can then be recorded in the field in analog or digital form and transmitted to a data processing center for further processing or can be further processed directly in the field. The further processing can involve, for example, writing source and receiver location data into each seismic record, filtering to remove undesired noise, gain correction, editing to remove undesirable traces, deconvolution, common depth point (CDP) sorting, corrections for normal moveout, corrections for statics, and the like.
The source-detector configurations for the generation and collection of seismic data include configurations for two-dimensional and for three-dimensional seismic prospecting. A configuration for two-dimensional seismic prospecting typically has the detectors arrayed along a line so that a detector location can be determined by the position of the detector in the line. A configuration for three-dimensional seismic prospecting typically has the detectors arranged in a configuration which would require, to determine the location of each detector, specifying two values, for example, an X and a Y coordinate, or longitude and latitude, etc.
Conventional two-dimensional seismic exploration techniques are relatively insensitive to error in the location of the seismic sources and receivers, hereafter location error. This relative insensitivity to location error results because seismic sources and receivers are, as indicated, typically located along a seismic line so that only locations along the line are required and these locations are relatively easy to determine using, for example, a surveying chain.
In three-dimensional seismic exploration techniques now in use and being developed, an accurate knowledge of the positions of the seismic sources and detectors used during the acquisition of the seismic data is a requirement. These locations are required for processing of the seismic signals, and further, as indicated, both X and Y coordinates (or latitude and longitude, etc.) of the source and detector locations must be known to properly process, for example, to stack and migrate and otherwise process the recorded seismic data. Location errors in the positions of the seismic sources and detectors can seriously degrade the quality of the resulting seismic images of the subsurface.
Location errors can occur or develop in a number of ways. Some possibilities include, for example, (1) surveying errors; (2) incorrect estimation of missing station locations after station flags are destroyed; (3) communication errors between shooter and observer concerning what source point is being initiated; (4) processing errors; and (5) bookkeeping errors anywhere along the surveyor-driller-shooter-observer-processor sequence.
Accordingly an object of the invention is a method for determining relative location errors of a seismic record.
A further object of the invention is a method for determining and correcting relative location errors of a seismic record.
Other objects of the invention are such methods applied to seismic records comprising data obtained from three-dimensional seismic prospecting.
Yet other objects are displays of seismic data in which relative location errors are displayed and/or corrected.
Further objects and advantages of the invention will be apparent to persons in the art from the following disclosure and the drawings.
Broadly, the invention comprises comparing a modeled event to an actual event representative of relative location to determine relative location error.
Thus, according to the invention is means and method for processing one or more seismic signals to determine relative location error comprising generating one or more first signals representative of a selected event representative of relative location according to a model of the selected event and determining relative location error by comparing one or more first signals with one or more second signals representative of the selected event as represented in the one or more seismic signals.
Further according to the invention is means and method for processing one or more seismic signals to determine relative location error comprising generating one or more first signals representative of a selected event in the seismic signal according to a model of the selected event, placing the thus generated first signals in or on the seismic signals, and displaying the resulting set of seismic signals in a seismographic display.
Further according to the invention is means and method for processing one or more seismic signals comprising generating for each of the one or more seismic signals a first signal representative of a selected event in the seismic signal according to a model of the selected event, comparing one or more of the first signals with one or more second signals representative of the selected event as represented in the one or more seismic signals, producing a signal representative of relative location error, and correcting the seismic signals according to the signal representative of relative location error to reduce relative location error distortion in seismic images produced using the seismic signals.
Other aspects of the invention will be appreciated by those in the art from the following detailed description and the drawings.
FIG. 1(A) is a schematic representation of a first configuration for seismic exploration.
FIG. 1(B) is a plot of data obtained from the configuration of FIG. 1(A).
FIG. 2(A) is a schematic representation of a second configuration for seismic exploration.
FIG. 2(B) is a plot of data obtained from the configuration of FIG. 2(A).
FIG. 3(A) is a schematic representation of a third configuration for seismic exploration.
FIG. 3(B) is a plot of data obtained from the configuration of FIG. 3(A).
FIG. 4 is a schematic representation according to one aspect of the invention of a first signal pattern and an event pattern showing isolated detector group error.
According to the invention, a seismic signal is processed to determine and/or display and/or correct relative location error. Each seismic signal is dependent upon source location and receiver location as well as upon subsurface propagation, reflection, refraction, and diffraction of the seismic signal. The location of each receiver can be defined relative to a source, or conversely the location of each source can be defined relative to the receivers, i.e., sources and receivers have "relative location". Sources and receivers also have what can be called absolute location, for example, defined by longitude and latitude, and the like. The method according to the invention is useful for determining relative location error; i.e., error in specifying the location of sources relative to receivers or vice versa. A consequence of the fact that the method is useful for determination of relative location error is that in practicing the method of the invention it is necessary to assume that the source location or the receiver location(s) for a given seismic signal is correct. As a practical matter, errors in source relative location in a configuration for seismic exploration can occur more frequently than errors in receiver relative location. Consequently, in the description of the illustrated embodiments herein it is assumed that receiver location is correct and source relative location error is determined. However, the principles herein described also apply when it is assumed source location is correct to determine receiver relative location error.
The seismic signal which can be processed using the method of the invention can be any seismic signal which contains an event representative of relative location. Practically speaking, it is preferable to process seismic signals according to the invention before such conventional processing as normal moveout corrections, common depth point sorting, and the like have been applied to the signals because such processing makes more difficult the determination, and correction, of relative location error even though such error determination and correction remain possible in principle. Preferably the seismic signal is a seismic signal such as is received in the field or recorded in the field following a seismic initiation. For purposes of describing the illustrated embodiment, the seismic signals are seismic records obtained in the field and processed in a processing center as a common-source record. However, direct processing in the field is also possible, as is processing of common receiver records, and many other variations which one skilled in the art can appreciate.
According to the invention, one or more first signals representative of a selected event according to a model is compared with one or more second signals representative of the selected event in the one or more seismic traces to determine relative location error. The selected event must be representative of relative location and can be any event in the seismic record for which a signal can be produced according to a model of the event. The model can be a physical model or a mathematical model. Preferably, the model is a mathematical model because such models are readily amenable to seismic data processing procedures.
According to the preferred embodiment, the event representative of relative location is a time event, specifically first arrival time. The principles are applicable to first arrival time of direct, reflected, refracted, and diffracted waves although in the preferred embodiment refracted waves are involved.
First arrival time can be approximated by the formula ##EQU1## where T is the first arrival time;
D is the source-receiver offset distance; i.e., the distance between the source and the respective receiver or detector; and
V is a velocity equal to the velocity of the seismic first arrival, for example, in the illustrated embodiment, refraction break. This velocity is assumed constant in the three-dimensional survey area.
Frequently, the value of V will be known from previous seismic exploration of the area. It is, however, also possible to determine V, for example, by including in the data, data derived from initiating a seismic source located in a line of detectors (conventional b 2-dimensional seismic prospecting) to give V for the remaining seismic exploration of the area whether 2-dimensional or 3-dimensional. A third way of determining V is by including V as the third unknown in a program to determine x-coordinates and y-coordinates of sources or receivers and solving by mathematical and computational techniques now available. The method by which V is determined, however, is not to be considered to limit the present invention which can be used with any suitable method for determining V.
Other formulas and models for approximating first arrival time which are representative of relative location can of course also be employed.
According to the invention, a first arrival time according to a model is compared to an actual first arrival time of one or more seismic signals to determine relative location error. In one illustrated embodiment, this comparison is carried out as follows.
As a part of processing seismic records for use, source and receiver locations can be input to a program which writes the appropriate source and detector, x, y coordinates, longitude, latitude or the like into the trace header for each seismic trace that was recorded. Using these coordinates, such a program can also write the distance between the source and the receiver ("source detector offset distance" or "D" in equation (1) above) into each trace header. Developing such programs is a routine part of seismic data processing and will not be further described. In the illustrated embodiment the traces preferably have such information in the trace header. Having such information in the trace header is not necessary for the practice of the invention but is one aspect of the invention which may be practiced in a number of ways readily devised by one skilled in seismic data processing.
According to the invention first signals are generated representative of first arrival time. In a preferred embodiment respective first signals are inserted into or superimposed upon each seismic signal at a time given according to the model of equation (1). To facilitate clarity in a seismographic display, the signals can be placed in the seismic signal by time-delaying each seismic record by a fixed amount, for example, 200 milliseconds (msec) to put a zone of zeroed signal at the start of each trace and by adding a first signal pulse to each seismic signal at a time according to the model of equation (1) described above. The method according to the invention can be profitably applied to a single seismic signal provided the user has confidence in the knowledge of subsurface velocity V. Preferably, however, the method is applied to more than one seismic signal to allow for variations in subsurface velocity which can occur. In a preferred embodiment, after a first signal is placed in each seismic signal, each common-source record can then be plotted to produce a seismographic display, using, for example, a conventional seismic plotter. The first signal pulses added to each seismic signal form a pattern when examined over many traces of a common-source record which is indicative of relative location error. In the absence of relative location error, the first signal pulses pattern will track the pattern formed by the first arrival pattern by a generally constant time difference approximately equal to the time delay, if any, that was applied to the seismic trace records, in the illustrated embodiment, 200 msec. If the first signal pulses pattern tracks the first arrival events pattern by a generally constant time difference, then no relative location error is indicated; and the described position of the source relative to the spread of receivers or detectors is probably correct. If the first signal pulse pattern does not track the first arrival pattern, then either the described source position or the described detector positions are indicated to be incorrect provided that the model, including, for example, velocity V determination in equation (1) above, is reasonably accurate. The failure to track is because the time of the first break event can be approximately described by the model of equation (1) above where D is the actual source-receiver offset distance.
FIG. 1, FIG. 2, and FIG. 3 illustrate this aspect of the invention through an example.
The data of FIGS. 1(B), 2(B), and 3(B) were recorded in a common-source record from a seismometer array as set forth in FIGS. 1(A), 2(A), and 3(A), respectively with 200 foot spacing between detector groups and 800 ft intervals between lines of detectors. The seismometers were Geospace Corporation Seismometers Model GSC-20D with each detector group comprising twelve seismometers laid in a linear array along the seismometer line. The recording system was a 120-channel DFS-V available from Texas Instruments. The source was dynamite. The retraction velocity V for the area prospected is known from previous exploration to be about 900 ft/sec.
FIG. 1A is a schematic representation of a configuration for three-dimensional seismic prospecting having four parallel east-west lines A, B, C, D, of detectors and having a seismic source at reference numeral 11. Each line or spread has 30 detectors as indicated by the numbers at the ends of each line so that a total of 120 seismic traces are recorded for each seismic initiation. FIG. 1(B) shows the first signal pulse pattern 15 and the first arrival pattern 16 wherein the first signal pulse pattern 15 tracks the first arrival pattern 16 by a generally constant offset of about 200 msec.
FIG. 2(A) is a schematic representation of an array as in FIG. 1(A) having the lines of detectors designated A', B', C', D'. The recorded source location is represented at reference numeral 21 whereas the actual source location as determined according to the invention is at reference numeral 22; i.e., there is source location error in the x dimension. FIG. 2(B) shows the first signal pulse pattern 25 and the first arrival pattern 26 for such an array. In this case, the error in the X-coordinate of the source has caused a shift in the x-dimension (detector group number) of the first signal pulse pattern relative to the first arrival pulse pattern. The amount of error in the x-dimension is equal to the number of traces which the first signal pulse pattern is shifted in the x-direction times the spacing between detectors. In this case, the error has shifted the pattern five (5) traces and assuming an average spacing of 200 ft between detectors the x-dimension error is 1000 ft. The determined error can be used to correct the seismic signals, for example, by inputting the x-dimension error to a program which will write the corrected x-dimension source position indicated by reference numeral 22 in the header for each trace of a common source record. The corrected records can then be further processed in any appropriate manner.
FIG. 3(A) is a schematic representation of an array as in FIG. 1(A) having the lines of detectors designated A", B", C", D". The recorded source location is at reference numeral 31 whereas the actual source location is at reference numeral 32; i.e., there is source location error in the y-dimension or time dimension. FIG. 3(B) shows the first signal pulse pattern 35 and the first arrival pattern 36 for the array of FIG. 3(A). The first signal pulse pattern indicates that the source is nearest the detector row consisting of receivers 61-90 whereas the first arrival signal pattern indicates that the source is nearest the detector row consisting of receivers 31-60. The x coordinate of the source is indicated to be correct since there is no shift in the x-direction of the first signal pulse pattern relative to the first arrival pattern. Y-coordinate error can be determined by measuring the time difference (Δt) between corresponding apexes on the first signal pulse pattern 35 and on the first arrival pattern 36. At the apexes, the x-coordinates of the source and the detector are the same, facilitating determination of y-coordinate error. In practice, determination of y-coordinate error is most easily accomplished by measuring the time difference (Δt) between the modeled time of first arrival and the recorded time of first arrival. In the event that the first pulse pattern has been shifted in time from the first arrival pattern by, for example, inserting a time delay at the start of each trace, then the time difference Δt of interest will be given by
when tdelay is the interval by which the traces are delayed, in the illustrated embodiment 200 msec. The y coordinate error is then given by
Δyerror =Δt V
where V is known as described above.
In the above-described example, the effect of relative source location error in the x-dimension and in the y-dimension are separable because the seismometer spreads are parallel to the x-axis. If the spreads are oriented skew to the x, y coordinate system relative location errors are still detectable but less easily separable since either x or y location error will cause both a shift and a distortion of the first pulse pattern relative to the first break pattern. This coupling of the two effects can be eliminated during seismic data processing by performing a rotation of coordinates from the x, y coordinates to a set of coordinates x', y' having x' oriented parallel to the seismometer spreads. Errors in x', y' of the location coordinates are then separately detectable and an inverse rotation back to x, y coordinates will give the error in x, y coordinates of the source point.
Further according to the invention relative location errors of the detectors in a common-source record, for example, are also determinable. According to the illustrated embodiment, of course, a constant error in the coordinates of an entire seismometer (detector) spread is indistinguishable from a source location error because the formula used in approximating first arrival time depends on the source-receiver offset distance; i.e., the difference of the source and receiver coordinates. Nevertheless, errors in the coordinates of a relatively small number of detectors are determinable as isolated departures from the tracking by the first signal pattern of the first arrival pattern. This situation is illustrated in FIG. 4 wherein reference numeral 41 refers to the first arrival pattern, reference numeral 42 refers to the first signal pulse pattern and reference numerals 43 refer to detectors having erroneous position descriptions.
As illustrated by the above description, relative location errors in, for example, a common source record, can be detected by the method of the invention. Further, according to the invention, the relative location errors can be corrected to within an accuracy tolerable for seismic exploration by analysis of first signal pulse pattern and first arrival pattern combined with a knowledge of the seismic array geometry and dimensions. The displays produced by practicing the method of the invention are also within the scope of the invention.
Further, according to the invention, however, is a method for processing seismic data in which no display need necessarily be generated before processing the corrected seismic data. According to this aspect of the invention, the recorded source and receiver location data are used to model the selected event. The modeled event is compared to the actual event in the seismic signal(s), an error signal representative of error in source location generated, and the generated error signal is used to correct the source and receiver location data. This can be readily accomplished by one skilled in the art by, for example, computer simulation of the method set forth above for determining and correcting location data errors. Thus, for example, the coordinates can be rotated as described above, if necessary. Errors in x- and y-coordinates can the be calculated as described and used to correct the location data, again with an inverse rotation of coordinates if necessary. Other methods, such as, for example, those involving the use of least square minimization can, of course, also be used and these techniques being in common use and familiar to those skilled in the art need not be further described.
The process of the present invention can be carried out using ananlog computing systems or digital computing systems. In the illustrated embodiment, a field recorder is used to record the seismic signals in the field and a conventional general purpose digital computer is utilized to carry out the process of the present invention on the recorded signals. One particular computing system which is suitable for use is International Business Machines IBM 370/168 with associated input/output devices and a particular plotter is a Geospace Model 3400 plotter. The apparatus of the invention comprises such analog computing systems and digital computing systems which have been configured or programmed to carry out the invention. Such configuration and programming as would be required to do this are routine to persons in the art who are familiarized with the invention and need not be further described.
It will be readily appreciated that the method and displays of the invention can be used to improve seismic data and to permit data to be used which would otherwise be omitted. Practice of the invention can produce, for example, improved determination of static corrections, improved determination of CDP stacking velocities, improved coherency of CDP stacking, consequently improved displays of migrated image of subsurface structure, and the like, and is easily implemented. Although the invention has been described in terms of preferred embodiments, however, the invention is not to be limited by the illusrated embodiments but by the claims appended hereto.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3214700 *||Mar 17, 1961||Oct 26, 1965||Trw Inc||Variable threshold signal detection system|
|US3284765 *||Sep 3, 1963||Nov 8, 1966||Mobil Oil Corp||Apparatus for establishing a common datum plane|
|US3500543 *||Oct 31, 1967||Mar 17, 1970||Petty Geophysical Eng Co||Method for editing seismic traces|
|US3569920 *||Jun 6, 1969||Mar 9, 1971||Edo Corp||Heading-sensing apparatus|
|US3864667 *||Sep 11, 1970||Feb 4, 1975||Continental Oil Co||Apparatus for surface wave parameter determination|
|US3924450 *||May 10, 1974||Dec 9, 1975||Hitachi Shipbuilding Eng Co||Device for measuring three dimensional coordinates of models|
|US3962671 *||Aug 18, 1975||Jun 8, 1976||The United States Of America As Represented By The Secretary Of The Navy||Calibration transducer|
|US3985024 *||Feb 28, 1975||Oct 12, 1976||Grumman Corporation||Acoustic emission process and system for improved flaw source location|
|US4240281 *||Nov 27, 1978||Dec 23, 1980||Mannesmann Aktiengesellschaft||Automatic self-checking of test equipment|
|US4323990 *||Feb 4, 1980||Apr 6, 1982||Mobil Oil Corporation||Seismic exploration system|
|1||*||Davis, James M., Interpretation of Velocity Spectra through an Adoptive Modeling Strategy, Geophysics, vol. 37, No. 6, 12/72, pp. 953 962.|
|2||Davis, James M., Interpretation of Velocity Spectra through an Adoptive Modeling Strategy, Geophysics, vol. 37, No. 6, 12/72, pp. 953-962.|
|3||*||Hilterman, Fred J., Three Dimensional Seismic Modeling, Geophysics, vol. 35, No. 6, 12/70, p. 1020 10 37.|
|4||Hilterman, Fred J., Three Dimensional Seismic Modeling, Geophysics, vol. 35, No. 6, 12/70, p. 1020-10-37.|
|5||*||J. A. Coffeen, Exploration Fundamentals, PPC Book, Tulsa, Oklahoma, 1978, pp. 51, 54 56.|
|6||J. A. Coffeen, Exploration Fundamentals, PPC Book, Tulsa, Oklahoma, 1978, pp. 51, 54-56.|
|7||*||Robert E. Sheriff, Encyclopedic Dictionary of Exploration Geophysics, Society of Exploration Geophysicists, Tulsa, Oklahoma, 1975, pp. 127, 128, 205, and 212.|
|8||*||Rumbold, J. G., Jackin, a New Approach to Calibrating, 39th National Fall Conference, 10/15 18/79, pp. 459 461.|
|9||Rumbold, J. G., Jackin, a New Approach to Calibrating, 39th National Fall Conference, 10/15-18/79, pp. 459-461.|
|10||*||Sheriff, Exploration Geophysics Encyclopedic Dictionary, 1/23/75, p. 143.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4698759 *||Mar 28, 1985||Oct 6, 1987||The United States Of America As Represented By The United States Department Of Energy||Process for structural geologic analysis of topography and point data|
|US4953142 *||Jan 6, 1989||Aug 28, 1990||Marathon Oil Company||Model-based depth processing of seismic data|
|US5157638 *||Jan 13, 1992||Oct 20, 1992||Conoco Inc.||Method of deriving statics corrections from common reflection point gathers|
|US6631343 *||Jan 8, 1998||Oct 7, 2003||Geotop Corporation||Method for reducing the calculation time of a numerical calculation for a computer implemented superposition model|
|EP0249527A2 *||Jun 1, 1987||Dec 16, 1987||Compagnie Generale De Geophysique||Method and apparatus for seismic data processing by correlation|
|EP0249527A3 *||Jun 1, 1987||Mar 15, 1989||Compagnie Generale De Geophysique||Method and apparatus for seismic data processing by correlation|
|EP0541265A2 *||Oct 19, 1992||May 12, 1993||Geco A.S.||Method of processing seismic data|
|EP0541265A3 *||Oct 19, 1992||Mar 9, 1994||Geco As||Title not available|
|U.S. Classification||367/73, 73/156|
|International Classification||G01V1/36, G01V1/28|
|Cooperative Classification||G01V1/362, G01V2210/53, G01V1/288|
|European Classification||G01V1/36B, G01V1/28D|
|Jan 21, 1986||AS||Assignment|
Owner name: AMOCO CORPORATION
Free format text: CHANGE OF NAME;ASSIGNOR:STANDARD OIL COMPANY;REEL/FRAME:004558/0872
Effective date: 19850423
Owner name: AMOCO CORPORATION,ILLINOIS
Free format text: CHANGE OF NAME;ASSIGNOR:STANDARD OIL COMPANY;REEL/FRAME:004558/0872
Effective date: 19850423
|Jan 19, 1988||CC||Certificate of correction|
|Sep 15, 1989||FPAY||Fee payment|
Year of fee payment: 4
|Jan 4, 1994||REMI||Maintenance fee reminder mailed|
|May 29, 1994||LAPS||Lapse for failure to pay maintenance fees|
|Aug 9, 1994||FP||Expired due to failure to pay maintenance fee|
Effective date: 19940529